1. Field of The Invention
This invention relates to new immobilized substrates and processes for their preparation and use in identifying and quantifying production and/or secretion of an enzyme specific to a cell type.
2. Description of Prior Art
A considerable body of research has centered on improving the early diagnosis of invasive and degenerative diseases such as cancer, arthritis and pulmonary emphysema. Standard methods for determining identities of pathogenic bacteria associated with infections are lengthy and often entail a wait of up to two days after taking cultures. These diseases involve synthesis and secretion by tissues of proteolytic enzymes (proteinases) that are capable of catalyzing the breakdown of proteinaceous materials. In cancer primarily connective tissues (collagen, elastin, etc.) which are hydrolyzed, and thus degraded, during the processes of tumor invasion and metastasis, or during the degradation of pulmonary alveolar membranes. In arthritis, articular cartilage surfaces are eroded enzymatically by the action of a variety of proteases collectively known as collagenases. A broad range of bacterial cells secrete other proteinases that are used to destroy antibodies produced and employed by the body to limit the invasion of microbes.
Uncontrolled destruction of intra- or extracellular proteins by proteinases is associated with many pathological conditions such as the breakdown of articular cartilage by elastase during rheumatoid arthritis, the destruction of pulmonary elastin by elastase during emphysema, and the activation of plasminogen during the invasion of healthy tissues by tumor cells. Proteolytic enzymes also play a role in rotavirus infectivity and are important for rotavirus propagation.
Proteolytic enzymes are involved in loss of growth regulation, invasiveness, metastasis, and formation of malignant tumors. Although tumor cells produce a wide variety of proteinases, the occurence of high levels of plaminogen activator (PA) is uniquely associated with malignancy. Elastase, a serine proteinase has been isolated and characterized from the human granulocyte (PMN). This enzyme is responsible for the release of proteoglycan matrix from intact articular cartilage and can degrade isolated aggregates or subunits of proteoglycan in solution. Elastase also stimulates cellular migration to the site of inflammation, can cleave intermolecular cross links of collagen, and can also cleave the helical portion of collagen III and IV. The latter action of elastase could be of the utmost importance in tissue damage. It is of great importance then to characterize the proteinase active in the acute response to inflammatory disease. Although similar proteinase activity is found in macrophages and in neutrophils, the major natural proteinase secreted by activated or inflammatory macrophages is plaminogen activator while elastase is most probably the key enzyme in neutrophil mediated proteolysis.
Simple synthetic substrates in which only one bond is susceptible to enzymatic hydrolysis often are used to assay serine proteinases. Most often, esters of L amino acids blocked at the N terminal are used. Ester substrates have the advantage over the corresponding amides in that Km values are lower and catalytic rate constants are higher. However, esters present difficulty in measuring the products of hydrolysis, i.e., the alcohol and the free amino acid. They also are not as well-selectively distinguished in the process of enzyme recognition. For example, Visser et al., 268 Biochim. Biophys. Acta 257 (1972) introduced the use of Boc-Alanyl-p-nitrophenyl ester as a substrate for elastase. However, it is also susceptible to hydrolysis by trypsin and chymotrypsin, is poorly water soluble and undergoes considerably spontaneous hydrolysis at pH 8.0. The acetyl-(alanine).sub.3 -methyl ester substrate that was proposed by Gertler et al., 48 Canad. J. Biochem. 384 (1970) overcomes most of these disadvantages, but its enzymatic hydrolysis must be monitored by time consuming potentiometric methods. The general applicability of esterolytic assays has been limited.
Studies of the action of a proteinase on small natural peptides of known sequences also have been useful in obtaining a broad characterization of the enzyme in question. The oligopeptides obtained, after proteolytic digestion under controlled conditions, are separated and identified.
For measurement of general proteolytic activity, the method of Kunitz, 22 J. Gen. Physiol. 429-446 (1939) is usually applied. This method uses casein as the substrate and a spectrophotometric method which depends upon differences in absorption between the products of hydrolysis and the substrate. The rate of change in absorption at a selected wavelength is proportional to the rate of hydrolysis.
Erlanger et al., 3 Biochemistry 346 (1961) describe the use of the p-nitroanilide of N benzylarginine (BAPNA) as a substrate for trypsin like proteinases. Colorimetric assays employing this substrate have two major or advantages over previous assays. First, the bond that is cleaved is an amide bond. Selectivity of proteolytic enzymes is much better for amide bonds than ester bonds. Second, the p-nitroanilide derivative is a chromogenic substrate. When it is cleaved by the enzyme, p-nitroaniline is released, which has a strong ultraviolet absorption at a different wavelength from that of the substrate. p-Nitroaniline is an aromatic amine with an extinction coefficient of 8,800 at 410 nm (Erlanger et al., supra). This compound, however, cannot be detected below 10.sup.-7 M unless it is converted by diazo coupling into a derivative having a stronger chromphore (Bieth et al., 53(2) Biochem. Biophys. Res. Comm. 383390 (1973)).
Fluorogenic substrates for proteinases have recently drawn considerable attention because they are capable of lowering the limit of detection obtainable from peptidyl p-nitroanilide substrates by as much as two orders of magnitude. The fluorogenic portion is generally an acylated aromatic amine. Upon hydrolysis, the spectroscopic properties of the leaving groups usually shift to longer wavelengths of absorption and/or emission, that are characteristic for the particular fluorophore.
Zimmerman et al., 70 Anal. Biochem. 258-262 (1976) describe the use of the fluorogenic substrate, 7 -glutaryl phenylalanylamido-4-methylcoumarin as a stable amide substrate for chymotrypsin. The sensitivity of this substrate is due to the fact that the leaving group, 7-amino-4 -methylcoumarin (AMC), is highly fluorescent. The amides are stable in solution and are in addition more closely related to the natural substrates. In a later study (Zimmerman et al., 78 Anal. Biochem, 47-51 (1977)), the preparation and use of fluorogenic amide using aminocoumarin as the leaving group was described for trypsin and elastase.
Castillo et al., 99 Anal. Biochem. 53-64 (1979) reported different substrates in which the leaving group was varied. The sequence used for the peptide was MeOSuc-Ala-Ala-Pro-Val-X, which had been shown to be highly reactive and relatively specific toward human leukocyte elastase (HLE). The X in the sequence represented 4-nitroanilide ( NA), thiobenzyl ester (-SBzl), 4-methyl-7-aminocoumarylamide (-AMC), or 1-methoxy-3-naphthylamide (-NNapOMe). The thiol benzyl ester substrate was shown to be the best of the series. Advantages of this substrate are its high kcatKm values and ease of synthesis. However, there is often interference from high concentrations of thiols present in samples such as cellular extracts or culture media. The peptidyl-AMC substrates are as sensitive as the thiol benzyl ester. They are, however, more difficult to synthesize and have lower kcat/Km values. The enzymatic cleavage does, in this case, involve a peptide linkage. However, for a rate assay, it had a lower sensitivity than either the thiobenzyl ester or peptidyl-AMC.
The following patents and reports also describe the use and application of solution chromogenic substrates: U.S. Pat. Nos. 3,884,896; 3,886,136; 4,016,042; 4,070,245 and 4,188,264; Plapinger et al 30 J Org Chem. 1781-1785 (1965) and Nachlas et al., 108 Arch. Biochem. Biophys., 266-274 (1964).
An assay for proteinases based on the fluorescent labeling of insoluble proteins (fibrin) or of soluble casein was described by Wiesner et al., 121 Anal. Biochem. 290-294 (1082). The fluorogenic reagent 2-methoxy-2,4-diphenyl-3(2H)-furanone (MDPF) was used to label the proteins. Fluorescence of the liberated peptide fluorophore conjugates resulting from enzymatic hydrolysis was measured in the supernatant after separation of the unreactive casein-fluorophore by acid precipitation. However, the enzymatic hydrolysis cannot be followed continuously during the time of incubation since interruption of the reaction is necessary for removal of the supernatant products.
Dipeptide derivatives of rhodamine were reported by Leytus et al., 215 Biochem. J. 253-260 (1983) as having a high degree of sensitivity and selectivity.
Synthetic chromogenic or fluorogenic substrates composed of peptidyl amides of aromatic amines have been widely used to detect and quantify proteinases and to define their amino acid specificities (Knight, Proteases in Mammalian Cells and Tissues, North Holland Pub. Co Amsterdam 1977)). Because of the sensitivity of fluorimetry, fluorogenic substrates are particularly suitable as probes of enzyme structure and mechanism. A number of problems, however, are generally encountered that limit the use of these agents.
First, the fluorescence spectra of the substrate and product often overlap to a significant degree, and it becomes necessary to use wavelengths longer than the excitation maximum to excite the product in order to avoid high levels of background emission from the excess unhydrolyzed substrate. This compromise decreases the fluorescence intensity used to estimate product, thereby lowering the sensitivity of the assay. For example, fluorogenic leaving groups such as .beta.-naphthylamine and 7-amino-4-methylcoumarin can be detected at nanomolar concentrations, yet this limit is not often achieved experimentally under assay conditions. If the spectrofluorimeter is adjusted to monitor the appearance of the product at its fluorescence maximum, high levels of background emission are produced by the large excess of substrate and it becomes exceedingly difficult to measure accurately the amount of amine liberated early in the course of substrate hydrolysis. Consequently, it is rare that both excitation and emission maxima can be selected for fluorescence studies of this type. Instead, a compromise must be reached between decreased background fluorescence at the expense of operating in the lower fluorescence spectral region of the product.
Second, the aromatic amines are usually hydrophobic and thus require organic cosolvents (e.g., dimethylformamide or dimethyl sulfoxide) to solubilize them. The presence of these solvents can produce unpredictable inhibitory or stimulatory effects on the enzyme. In addition, such solvents are inappropriate for the in situ assay of proteinases in cell culture.
Third, the currently available chromogenic or fluorogenic groups are not readily amenable to derivatization. A leaving group of the substrate should be readily derivatized into series of congeners which allow studies of the steric and electronic requirements of the active site of the enzyme. Moreover, derivatization of this sort permits the possibility of immobilizing the fluorogenic moiety on solid supports via covalent bond formation.
Fourth, these compounds are not conveniently used for the assay of single cells. Several laboratories have explored the application of chromogenic peptides as histochemical probes for revealing the presence of peptidase activity in cells (Dolbeare et al., 27(11) J. Histochem. Cytochem 1493-1495 (1979); and Sannes et al., 27 J. Histochem. Cytochem 1496-1497 (1979)); and tissue sections (Blasini et al., 13 Thrombosis Rev. 585-590 (1978); and Grabske et al., (11) J. Histochem. Cytochem. 1505-1508 (1979)). Although these substrates are selective reagents for the quantitation of various proteinases, a major shortcoming is that they are only practical for the assay of proteinases released into the extracellular medium by a population of cells. Another disadvantage of these cytochemical methods is that the low concentration of chromopore that is liberated by cellular proteinases is difficult to detect, and hence, it must often be visualized by conversion, using an azocoupling reaction, into derivatives having larger extinction coefficients. Unfortunately, the vigorous conditions needed to affect this transformation preclude its use with cells and tissues where one wishes to maintain viability.
Recently, a fluorescent proteinase transition state analog-inhibitor, dansyl-L-arginal (DansArgH) was introduced as a selective probe for cysteine and serine type proteinases in a fibrosarcoma tumor cell line (Kozlowski et al., 81 Proc. Nat'l. Acad. Sci. 1135-1139 (1984)). However, the conditions used to quantify and to visualize enzyme activity also prevented the maintenance of cell viability.
Fifth, a drawback of low molecular weight substrates is that they diffuse from the site of reaction, even if entrapped within an agarose or gelatin matrix. This limits their application as probes of proteinase activity in single-cells. As a result, only very brief exposure periods can be employed if enzyme activity is to be localized at an individual cell or discrete tissue region. These restrictions diminish the sensitivity of such methods and confine application to cells which secrete relatively large amounts of proteinase. Similar limitations also are encountered when radiolabeled substrates dried onto the surface of microtiter wells are used as an assay for proteolytic activity (Varani et al., 107 Anal. Biochem. 377-384 (1980)). Besides the usual disadvantage of handling radioactive materials, one cannot localize the regions of proteolytic activity since the radioactive product of the reaction is released into the supernatant fluid.
Sixth, a disadvantage of currently used proteinase detection methods concerns quantitative. Individual cells that secrete specific proteinases have been identified by techniques that involve the formation of a pericellular lysis zone in an opalescent agar matrix containing a natural substrate such as fibrin or casein (Jones et al., 5 Cell. 323-329, (1975)). Although this type of assay is sensitive, it is not quantitative because the relationship between the size of the lysis zone and the amount of enzyme in question is not known. Moreover, the agar matrix used in the system is not physiological and therefore some cells do not survive this treatment. Also, this technique can only be used for cells that grow in soft agar and it is capable of detecting only those enzymes that digest the macromolecular substrates. ##STR1## has been introduced (Bryness et al., 116 Anal. Biochem. 408-413 (1981)) as an exceptionally useful fluorogenic group for synthetic substrates because: (a) it is approximately the same size as, .beta.-naphthylamine, ##STR2## a fluorogenic moiety commonly used in proteinase substrates; (b) after alkylation of the ring nitrogen, it has a similar distribution of charges as the p-nitroaniline leaving group, ##STR3## (c) acy1aminoquinolines are known to fluoresce only weakly in the bluish-white region of the spectrum, whereas the free amine emits intensely yellowish green light (the part of the visual spectrum in which the eye is most sensitive); (d) the appearance of 6-AQ can be measured fluorometrically at its excitation and emission maxima, while at these wavelengths the substrate is essentially nonfluorescent (Brynes et al., supra, 1981); (e) the wavelength of maximal excitation of 6-AQ is sufficiently low in energy that chromophores of proteins and nucleic acids are not excited; (f) the amino group of 6-AQ can undergo acylation by peptides in high yield because it is more basic than the resonance-deactivated amino groups of other frequently used chromophores; and (g) the ring nitrogen of the quinolines is readily quaternized by a variety of alkylating agents, generating highly water soluble substrates.